Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A computer-readable storage medium storing computer-executable instructions for controlling a computing device to provide a graphical user interface for analyzing an entity graph having vertices and edges, at least some of the vertices being entity vertices representing an entity of an entity type, each entity having an entity identifier, each edge having an edge type, the instructions comprising: a generate overall characteristics report component that receives from a user a selection of one or more characteristics relating to the entity graph; generates an overall characteristics report for the entity graph that includes the selected characteristics; and presents the overall characteristics report to the user; and a generate entity type graph component that receives from a user a specification of a node filter and a link filter; generates an entity type graph for the entity graph, the entity type graph having nodes and links, each node representing an entity type and satisfying the node filter, each link connecting nodes to indicate that the entity graph includes one or more edges connecting entity vertices with the entity types of the connected nodes and satisfying the link filter; and presents the entity type graph to the user.
This invention relates to computer systems and graphical user interfaces for analyzing complex data structures. Specifically, it addresses the challenge of visualizing and understanding relationships within large entity graphs. The system provides a computer-readable storage medium containing instructions to control a computing device. This device generates a graphical user interface for analyzing an entity graph. An entity graph consists of vertices and edges. Some vertices represent specific entities, each with a unique identifier and belonging to an entity type. Edges represent relationships between entities and have a defined edge type. The system includes two primary components. The first, a generate overall characteristics report component, allows a user to select specific characteristics of the entire entity graph. It then generates and displays a report summarizing these selected characteristics. The second component, a generate entity type graph component, enables users to define filters for nodes and links. It then constructs and presents an entity type graph. In this entity type graph, nodes represent entity types that meet the user's node filter criteria. Links connect these nodes, indicating the presence of one or more edges in the original entity graph between entities of those types, provided these edges satisfy the user's link filter criteria. This allows for a high-level overview of relationships between different types of entities.
2. The computer-readable storage medium of claim 1 wherein the size of a node in the entity type graph is dependent on the number of entity vertices in the entity graph with the entity type of that node.
This invention relates to data visualization systems that use entity type graphs to represent relationships between entities. The problem addressed is the lack of intuitive visual representation of entity types in large datasets, making it difficult for users to quickly understand the distribution and significance of different entity types. The system generates an entity type graph where nodes represent entity types and edges represent relationships between them. The size of each node in the entity type graph is dynamically adjusted based on the number of entity vertices in the underlying entity graph that share the same entity type. This means that nodes representing more frequently occurring entity types will appear larger, providing an immediate visual indication of their prevalence. The graph may also include additional visual indicators such as color or edge thickness to further enhance the representation of relationships and importance. The system processes input data to identify entities and their types, constructs the entity graph with vertices and edges, and then generates the entity type graph with sized nodes. This visualization helps users quickly assess the distribution of entity types and their interconnections, improving data analysis and decision-making in domains like social networks, biological data, or knowledge graphs.
3. The computer-readable storage medium of claim 1 wherein the width of a link in the entity type graph is dependent on the number of edges in the entity graph that connect vertices with the entity types of the nodes connected by the link.
This invention relates to data visualization in entity type graphs, specifically improving the representation of relationships between entity types. The problem addressed is the lack of intuitive visual cues in existing entity type graphs to convey the strength or significance of relationships between different entity types. Current methods often use simple lines or connections without conveying the underlying data volume or importance of these relationships. The invention enhances entity type graphs by dynamically adjusting the width of links between nodes based on the number of edges in an associated entity graph that connect vertices with the corresponding entity types. In other words, if two entity types are frequently connected in the underlying data, the link between them in the entity type graph will appear thicker, providing a visual indication of the relationship's strength. This approach leverages an entity graph, which consists of vertices representing entities and edges representing relationships between those entities, to inform the visualization in the entity type graph. The entity type graph itself is a higher-level abstraction where nodes represent entity types rather than individual entities, and links represent possible relationships between those types. By making link width proportional to the number of connecting edges in the entity graph, the visualization becomes more informative, allowing users to quickly identify prominent relationships. This method improves data analysis by making complex relationship patterns more accessible and interpretable.
4. A method performed by a computer system for providing a graphical user interface for analyzing an entity graph having vertices and edges, at least some of the vertices being entity vertices representing an entity of an entity type, each entity having an entity identifier, each edge having an edge type, the method comprising: receiving from a user a specification of a node filter and a link filter; generating an entity type graph for the entity graph, the entity type graph having nodes and links, each node representing an entity type and satisfying the node filter, each link connecting nodes to indicate that the entity graph includes one or more edges connecting entity vertices with the entity types of the connected nodes and satisfying the link filter; and presenting the entity type graph to the user.
This invention relates to a computer-implemented method for analyzing entity graphs, which are data structures representing relationships between entities. The problem addressed is the complexity of visualizing and interpreting large-scale entity graphs, where individual entities and their connections can be overwhelming for users to analyze directly. The solution involves abstracting the detailed entity graph into a higher-level representation focused on entity types and their relationships, making the data more manageable and interpretable. The method operates by first receiving user-specified filters for nodes and links. These filters define criteria for selecting relevant entity types and the relationships between them. The system then generates an entity type graph, where nodes represent entity types that meet the node filter criteria, and links between nodes indicate that the original entity graph contains edges connecting entities of those types that satisfy the link filter. This abstraction simplifies the visualization by focusing on the structural patterns of entity types rather than individual entities. The resulting entity type graph is presented to the user, providing a clearer overview of the relationships within the data. This approach is particularly useful in applications like fraud detection, social network analysis, or knowledge graph exploration, where understanding high-level patterns is critical.
5. A computing device for generating an entity type graph, the computing device comprising: a computer-readable storage medium storing computer-executable instructions for controlling the computing device to: access an entity graph having vertices and edges, at least some of the vertices being entity vertices representing an entity of an entity type, each edge having an edge type; access a specification of a node filter and a link filter; for each entity type of an entity of the entity graph that satisfies the node filter, add a node to the entity type graph; and for at least some of the pairs of nodes of the entity type graph, add to the entity type graph a link connecting the nodes of the pair when the entity graph includes one or more edges connecting entity vertices with the entity types of the connected nodes and satisfying the link filter; and a processor for executing the computer-executable instruction stored in the computer-readable storage medium.
The invention relates to a computing device for generating an entity type graphs from an existing entity graph. The problem addressed is the need to extract and visualize specific entity types and their relationships from a larger, more complex entity graph. The computing device includes a storage medium storing instructions and a processor to execute those instructions. The instructions enable the device to access an entity graph composed of vertices and edges, where vertices represent entities of various types and edges represent relationships between those entities, each with an associated edge type. The device also accesses a specification of a node filter and a link filter. The node filter determines which entity types from the original graph are included in the new entity type graph, while the link filter determines which relationships between those entities are preserved. The device processes the original graph to create the entity type graph by adding nodes for each entity type that satisfies the node filter. For pairs of nodes in the entity type graph, the device adds links between them if the original graph contains edges connecting entities of those types that satisfy the link filter. This approach allows for the creation of simplified, focused graphs that highlight specific entity types and their relationships, making the data more interpretable and useful for analysis.
Unknown
May 19, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.